All,
As there is considerable interest regarding having a gigabit version of Protocol 2 firmware available to try on Hermes, Hermes 10, and Hermes 100 platforms (beta test) I have posted the current 1000BaseT beta test firmware version for Hermes onto the repository for those interested in trying it:
https://github.com/TAPR/OpenHPSDR-Firmware
This version of Protocol 2 firmware runs satisfactorily on my standalone Hermes board at all data sampling rates through 1.536Msps with the only modification made to my board being a short across F3 (bottom of the Hermes board). Note that I am not recommending that everyone short F3 on Hermes, I am simply stating what I did to mine upon receiving it some years ago; long before Protocol 2 came into existence. This firmware should be considered as beta test firmware, as Protocol 2 is not officially released at this time, and is provided simply to allow interested users the ability to try it on their boards if they desire to do so.
While PureSignal appears to function on my board when using this firmware it does not seem to update the PureSignal display during Tx mode, probably indicating that the timing constraints are not yet optimum within this particular firmware version; behavior on other Hermes boards may differ from mine. It is my feeling that this initial gigabit version may nevertheless find usefulness for some Hermes users to test with even if it is not yet an ideal version for every Hermes user.
For those not wishing to try this gigabit version but would like to try a Protocol 2 version of firmware, the Hermes_Protocol_2_v10.3 firmware previously posted Protocol 2 firmware version for Hermes is available on the same repository page which runs using a 100BaseT ethernet connection instead. It also should be considered to be beta test firmware, as Protocol 2 has not been officially released yet, of course.
73, Joe K5SO
Hermes Protocol 2 Gigabit firmware v10.4 available for beta testing
Forum rules
Until such time as the New Protocol firmware goes into general release, all discussion will be concentrated here.
Until such time as the New Protocol firmware goes into general release, all discussion will be concentrated here.
Re: Hermes Protocol 2 Gigabit firmware v10.4 available for beta testing
Thank you Joe! Also, thank you for clarifying that F3 is the resettable fuse that we should be aware of. Am I correct in assuming that as you increase the sampling rate beyond 192k that the current demand climbs and F3 heats up causing it to be more resistive with a resulting voltage drop to the FPGA? I'm just curious if users who try the 1000T firmware can run it at 48/96/192 without triggering that scenario.
73,
Rob W1AEX
73,
Rob W1AEX
"One thing I am certain of is that there is too much certainty in the world."
- Tony EI7BMB
- Posts: 653
- Joined: Sun Apr 09, 2017 2:31 pm
- Location: Dublin
- Contact:
Re: Hermes Protocol 2 Gigabit firmware v10.4 available for beta testing
Hi Rob,
I assume so but I'm no expert there. I purchased my Hermes from a TAPR run and as I recall it came to me with F3 shorted already. I have never used the board without F3 shorted so I don't know how it behaves without the fuse shorted.
73, Joe K5SO
I assume so but I'm no expert there. I purchased my Hermes from a TAPR run and as I recall it came to me with F3 shorted already. I have never used the board without F3 shorted so I don't know how it behaves without the fuse shorted.
73, Joe K5SO
Re: Hermes Protocol 2 Gigabit firmware v10.4 available for beta testing
Joe,
Thank you for the background and also for the 1000T release!
73,
Rob
Thank you for the background and also for the 1000T release!
73,
Rob
"One thing I am certain of is that there is too much certainty in the world."
Re: Hermes Protocol 2 Gigabit firmware v10.4 available for beta testing
Joe,
Just a follow-up with the Hermes 1000T v1.04 firmware. I loaded it into my ANAN-100 and it ran fine. It never went beyond 2.20% (1Gbps) of network utilization while running a 192k sampling rate on RX-1. Of course being adventurous, I stepped through all the sampling rates right up to 1536k just to see what it would do. It ran them all, however, I did not linger at 1536k for much longer than around 30 seconds figuring F3 was possibly running in the transition zone and heating up. While running 1536k the network utilization sat steadily at 8.94% or 89.4 Mbps according to the Windows Task Manager - Networking monitor.
I could not get Pure Signal to work at 192k/96k/48k but otherwise it transmitted fine with Pure Signal disabled. For some reason, I can no longer get Pure Signal to correct with the 100T 1.03 firmware either which is a mystery to me because it worked fine the first time I ran it with the 1.03 firmware. I'll have to look at what I might have changed in my THETIS settings.
Thanks again for the 100T and 1000T Hermes Protocol 2 firmware versions.
73,
Rob W1AEX
![Image](http://www.w1aex.com/archive/Hermes_1.04_1536k.jpg)
Just a follow-up with the Hermes 1000T v1.04 firmware. I loaded it into my ANAN-100 and it ran fine. It never went beyond 2.20% (1Gbps) of network utilization while running a 192k sampling rate on RX-1. Of course being adventurous, I stepped through all the sampling rates right up to 1536k just to see what it would do. It ran them all, however, I did not linger at 1536k for much longer than around 30 seconds figuring F3 was possibly running in the transition zone and heating up. While running 1536k the network utilization sat steadily at 8.94% or 89.4 Mbps according to the Windows Task Manager - Networking monitor.
I could not get Pure Signal to work at 192k/96k/48k but otherwise it transmitted fine with Pure Signal disabled. For some reason, I can no longer get Pure Signal to correct with the 100T 1.03 firmware either which is a mystery to me because it worked fine the first time I ran it with the 1.03 firmware. I'll have to look at what I might have changed in my THETIS settings.
Thanks again for the 100T and 1000T Hermes Protocol 2 firmware versions.
73,
Rob W1AEX
![Image](http://www.w1aex.com/archive/Hermes_1.04_1536k.jpg)
"One thing I am certain of is that there is too much certainty in the world."
Re: Hermes Protocol 2 Gigabit firmware v10.4 available for beta testing
Hi Rob,
Thanks for the tests and reports. I also am able to reproduce here your results of PureSignal not working with Hermes_Protocol_2_v10.3 (100T) after testing Hermes_Protocol_2_v10.4 (1000T). To cure that I had to physically throw out the database.xml file and run Hermes_Protocol_2_v10.3. Simply resetting the database using the Thetis button did not allow recovery. I also disabled the "Reuse Last IP Address" and "Network Watchdog" options on the Thetis > Setup > H/W Select menu options; I don't know if that was necessary or not. I'm not familiar with the internal workings of Thetis with respect to the network operations/selections to know about how that affects things internally in Thetis.
However, and interesting to me, is that starting Thetis using Hermes_Protocol_2_v10.4 (1000T) with an absent database.xml file (i.e., allowing Thetis to construct a new one for me) I am able to get PureSignal to run and update if I disable the "auto-attenuate" button on the PureSignal advanced menu and manually select the attenuation level using the Thetis S-ATT control on the main Thetis display.
I suspect, as I mentioned earlier, that this is due to a slight mis-timing in the firmware somehow, but it doesn't explain to me why v10.3 doesn't work properly after v10.4 is run. Or if it does explain it somehow, I don't see why it does, at least. Perhaps this is a clue that I'm simply not understanding yet.
73, Joe K5SO
Thanks for the tests and reports. I also am able to reproduce here your results of PureSignal not working with Hermes_Protocol_2_v10.3 (100T) after testing Hermes_Protocol_2_v10.4 (1000T). To cure that I had to physically throw out the database.xml file and run Hermes_Protocol_2_v10.3. Simply resetting the database using the Thetis button did not allow recovery. I also disabled the "Reuse Last IP Address" and "Network Watchdog" options on the Thetis > Setup > H/W Select menu options; I don't know if that was necessary or not. I'm not familiar with the internal workings of Thetis with respect to the network operations/selections to know about how that affects things internally in Thetis.
However, and interesting to me, is that starting Thetis using Hermes_Protocol_2_v10.4 (1000T) with an absent database.xml file (i.e., allowing Thetis to construct a new one for me) I am able to get PureSignal to run and update if I disable the "auto-attenuate" button on the PureSignal advanced menu and manually select the attenuation level using the Thetis S-ATT control on the main Thetis display.
I suspect, as I mentioned earlier, that this is due to a slight mis-timing in the firmware somehow, but it doesn't explain to me why v10.3 doesn't work properly after v10.4 is run. Or if it does explain it somehow, I don't see why it does, at least. Perhaps this is a clue that I'm simply not understanding yet.
73, Joe K5SO
Re: Hermes Protocol 2 Gigabit firmware v10.4 available for beta testing
Joe,
Thanks for sharing what you have found. I have to step away for most of tomorrow, but when I have a chance I will see if I can duplicate your absent database.xml phenomenon with auto-attenuate disabled and manual adjustment of the S-ATT level.
73,
Rob
Thanks for sharing what you have found. I have to step away for most of tomorrow, but when I have a chance I will see if I can duplicate your absent database.xml phenomenon with auto-attenuate disabled and manual adjustment of the S-ATT level.
73,
Rob
"One thing I am certain of is that there is too much certainty in the world."
Re: Hermes Protocol 2 Gigabit firmware v10.4 available for beta testing
Sorry I slipped a cog in my precious post I think I may have referred to the new firmware as 10.8, not 10.3 as I should.
Re: Hermes Protocol 2 Gigabit firmware v10.4 available for beta testing
Second attempt to post.
I have installed this firmware on my Apache labs supplied Hermes board that I have been using successfully with Firmware 3.3 and PowerSDR.
Using Thetis 2.6.9 with the 10.4 most things seem to work. I have not tested Pure Signal However I have run into a a show stopper in my station. The control lined from the Hermes J16, set in the OC Control setup tab do not appear to be functional. I use them to set the LP filters in my Juma 1000 linear. The lines worked correctly in my previous PowerSDR setup. I don't know if this is a Firmware or Thetis problem. Or maybe I missed a Setup enable somewhere.
Just a note that may have been posted elsewhere. I had to use the HPSDR Bootloader for 10.4 on my board. Using the HPSDR Programmer it crashed while writing. It probably overwrote the code Programmer was using.
73, Dick
W0QM, G3FNL
I have installed this firmware on my Apache labs supplied Hermes board that I have been using successfully with Firmware 3.3 and PowerSDR.
Using Thetis 2.6.9 with the 10.4 most things seem to work. I have not tested Pure Signal However I have run into a a show stopper in my station. The control lined from the Hermes J16, set in the OC Control setup tab do not appear to be functional. I use them to set the LP filters in my Juma 1000 linear. The lines worked correctly in my previous PowerSDR setup. I don't know if this is a Firmware or Thetis problem. Or maybe I missed a Setup enable somewhere.
Just a note that may have been posted elsewhere. I had to use the HPSDR Bootloader for 10.4 on my board. Using the HPSDR Programmer it crashed while writing. It probably overwrote the code Programmer was using.
73, Dick
W0QM, G3FNL